banners
beforecontenttitle

The PREMOVE Dataset

Después del título del contenido
Antes del cuerpo del contenido
Trozos html editables
Trozos html editables

A multimodal dataset recorded from a sensorized vehicle in the Teatinos University Campus and two adjacent residential neighborhoods of Málaga, Spain. The dataset was designed to capture a wide variety of traffic participants including cars, pedestrians, motorcycles, bicycles, buses, and trucks operating in real and dynamic urban environments. In particular, six data sequences were recorded under different illumination conditions—daytime, sunset, and night—covering a total distance of 21,87 km and 61 minutes of driving.

The sensing platform integrates a redundant and complementary sensor suite designed to capture multimodal data around the frontal and lateral sides of the vehicle, which includes overlapping stereo RGB (Red-Green-Blue) and thermal cameras, three IMU (Inertial Measurement Unit) sensors, a dual 3D (three-dimensional) LiDAR (Light Detection and Ranging) stacked configuration, and a dual GNSS (Global Navigation Satellite System) with RTK (Real-Time Kinematic) differential corrections. 

   

Data is provided in human-readable format, complemented by the original rosbag. In addition, a collection of open-source Python tools is provided to ease data extraction, processing and synchronization.

The current state of the art shows that this dataset fills a gap in autonomous driving research by providing synchronized RGB–thermal–LiDAR data with overlapping frontal and lateral fields of view. Potential applications of the dataset include sensor fusion, perception, tracking, RGB-thermal domain adaptation and SLAM (Simultaneous Localization and Mapping).

This work has been performed in the frame of the project PREMOVE: Movement Prediction of Traffic Participants for the Safe Integration of Self-Driving Cars in Urban Areas" (ProyExcel_00684), funded by the Andalusian Regional Government (Junta de Andalucía) of Spain.

 

DATASETS

DESCRIPTION PATH DATASET LINKS VIDEO PREVIEW
30/10/2025 10:18:50 (UTC)
  • Teatinos Campus (Expansion)
  • Length: 1940m, 4min 19s 
  • Light conditions: Day 

 

 

 

RAW+BAG:

ampliacion-day-2025-10-30_files.zip 

Demo. Length: 40 s. Size: 3.20 GB

30/10/2025 10:21:26 (UTC)
  • Teatinos neighborhood
  • Length: 4340m, 12min 10s 
  • Light conditions: Day 

RAW+BAG:

teatinos-day-2025-10-30_files.zip  

Demo. Length: 50 s. Size 4.97 GB

10/11/2025 17:01:33 (UTC)
  • Teatinos Campus (Expansion)
  • Length: 3380m, 7min 40s
  • Light conditions: Sunset 

 

 

 

 

 

RAW+BAG:

ampliacion-sunset-2025-11-10_files.zip

Demo. Length: 40 s. Size: 2.91 GB

10/11/2025 17:09:13 (UTC)
  • Teatinos neighborhood
  • Length: 7360m, 23min 25s 
  • Light conditions: Sunset 

RAW+BAG:

teatinos-sunset-2025-11-10_files.zip 

Demo. Length: 40 s. Size: 2.45 GB

10/11/2025 17:38:25 (UTC)
  • Teatinos Campus (Expansion)
  • Length: 3060 m, 7min 57s 
  • Light conditions: Night 

RAW+BAG:

ampliacion-night-2025-11-10_files.zip 

Demo. Length: 60 s. Size: 3.80 GB

 

10/11/2025 17:32:38 (UTC)
  • Teatinos neighborhood
  • Length: 1790m, 5min 47s 
  • Light conditions: Night 

 

RAW+BAG:

teatinos-night_2025-11-10_files.zip 

Demo. Length: 40 s. Size: 2.93 GB

 

SUPPLEMENTARY MATERIAL

Github repository: Set of tools to facilitate data extraction, conversion, and processing from the original PREMOVE dataset recordings. Please check "README.md" file for installation instructions and usage.


Después de cuerpo del contenido